Digraph arc list implementation

I'm trying to implement graph as the Arc List, and while this implementation works efficiency is horrible, anything i missed that's making it so slow? Time was measured as the average of looking for arcs from/to each node.

struct Arc 
{
    int start;
    int end;
    Arc(int start,int end)
      : start(start),
        end(end)
    { }
};

typedef vector<Arc> ArcList;

class AListGraph
{
public:
    AListGraph(IMatrix* in); //Fills the data from Incidence Matrix
    bool IsE(int va,int vb); //checks if arc exists a->b
    int CountEdges(); //counts all the arcs
    int CountNext(int v); //counts all ou开发者_如何学JAVAtgoing arcs from v
    int CountPrev(int v); //counts all arcs incoming to v

private:
    ArcList L;
    int VCount;
};

//Cut out constructor for clarity

int AListGraph::CountEdges() 
{
    return L.size();
}

int AListGraph::CountNext(int v)
{
    int result=0;
    for(ArcList::iterator it =L.begin();it!=L.end();it++)
    {
        if(it->end==v)result++;
    }
    return result;
}

int AListGraph::CountPrev(int v)
{
    int result=0;
    for(ArcList::iterator it =L.begin();it!=L.end();it++)
    {
        if(it->start==v)result++;
    }
    return result;
}

Well, your implemention is worst possible : in order to find the in/out edges you have go across the whole graph.

Do you really need an arc list? Usually an adjacency list is much more efficient.

A naïve implementation of an adjacency list would be to keep an vector > arcs where the size of the arc is the number of nodes.

Given an node u, arcs[u] gives you all the out edges. You can figure out how to optimize it for in edges also.